摘要
针对混流装配线的多目标调度优化问题,提出了一种疫苗协同进化的多目标免疫克隆选择优化算法.设计了疫苗种群及其相关操作,使其跟抗体种群相互影响并协同进化,提高了算法的性能;针对调度优化问题的离散性,选择同时从抗体的基因型和表现型评价抗体亲和度;依据抗体质量和进化代数,设计了自适应变异率;在每次迭代过程中,通过多次局部寻优加快算法收敛速度.最后通过两组实例仿真,与另3种多目标优化算法进行比较,结果证明该算法可得到更好的计算结果.
In order to solve the scheduling optimization problem in mixed-model assembly lines, a multi-objective vaccine eoevolution elonal selection algorithm is proposed, and the vaccine population and the corresponding popu- lation operations are designed to interact and coevolve with the antibody population, thus greatly improving the per- formance of the algorithm. Then, according to the discrete feature of the scheduling optimization problem, the anti- body affinity is evaluated from the phenotype and the genotype. Moreover, according to the antibody quality and the evolutionary generations, the adaptive mutation rate is designed, and multiple local optimizations are executed in each iteration process to improve the convergence rate of the algorithm. The results of two series of experiments show that, as compared with other three multi-objective optimization algorithms, the proposed algorithm is of high efficiency and superiority.
出处
《华南理工大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2010年第3期133-137,142,共6页
Journal of South China University of Technology(Natural Science Edition)
基金
霍英东教育基金会青年教师基金资助项目(111056)
江苏省重大科技成果转化专项资金项目(BA2007034)
江苏省高校科技成果产业化推进项目(JH07-005)
教育部"新世纪优秀人才支持计划"资助项目(NCET080703)
关键词
混流装配线
多目标优化
疫苗
协同进化
克隆选择
mixed-model assembly line
multi-objective optimization
vaccine
coevolution
clonal selection
